Occasional write-ups about home servers, storage and small networks. No comments, no tracking.
The array had been degraded for eleven days before I noticed, which says more about my monitoring than about the disk. This is what the rebuild looked like, how long it actually took, and the three checks I added afterwards so the next failure is loud instead of quiet.
I replaced a clever setup with a dull one and everything got better. Notes on why I stopped running services on the gateway, and what I moved them to instead.
Averages hide everything interesting. A short tour of percentiles, why the ninety-ninth is the one that wakes you up at night, and a small script that plots them from ordinary ping output.
| Date | Change | Notes | Size |
|---|---|---|---|
| 2026-08-07 | Backup rebuild write-up | added | 18 KB |
| 2026-07-22 | Router post | added | 11 KB |
| 2026-07-14 | Uses page | updated | 6 KB |
| 2026-07-03 | Latency post | added | 24 KB |
| 2026-06-19 | Feed switched to Atom | changed | — |